Anal. Chem. 2007, 79, 8144-8153
Development and Application of C60-Fullerene Bound Silica for Solid-Phase Extraction of Biomolecules Rainer M. Vallant, Zoltan Szabo, Stefan Bachmann, Rania Bakry,* Muhammad Najam-ul-Haq, Matthias Rainer, Nico Heigl, Christine Petter, Christian W. Huck, and Gu1 nther K. Bonn
Institute of Analytical Chemistry and Radiochemistry, Leopold-Franzens University, Innrain 52a, 6020 Innsbruck, Austria
Sample pretreatment is the most important procedure to remove the matrix for interfacing with mass spectrometry (MS). Additionally, for the samples with low concentration, the process of preconcentration is required before MS analysis. We have newly developed a solid-phase extraction stationary phase based on C60-fullerene covalently bound to silica for purification of biomolecules of different characteristics. Silica particles of different porosity are modified with aminopropyl linker and then covalently bound to C60-fullerenoacetic acid or C60epoxyfullerenes. The developed materials have been successfully applied as an alternative to commercially available reversed-phase materials for solid-phase extraction. C60-fullerene silica is able to retain small and hydrophilic molecules like phosphopeptides, which can be easily lost by reversed-phase sorbents. The novel materials are applied for desalting and preconcentration of proteins and peptides, especially phosphopeptides. In addition, the C60-fullerene silica is applied for the solid-phase extraction of selected flavonoids with recoveries of ∼99%. The recoveries are compared with the commercially available solid-phase extraction materials. Mass spectrometry is a powerful instrumental technique for protein identification and characterization. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDITOF-MS) is widely used for the analysis of proteins and peptide mixtures. In proteomics, this method is frequently used for protein identification based on the comparison of experimentally measured molecular masses of enzyme-derived peptides with theoretical digests of primary protein structures obtained from databases.1-3 On the other hand, many samples for MALDI-TOF-MS mainly of biological origin are well-known to be complex mixtures of proteins with a wide range of molecular masses, salts, and other compounds. These components have a strong influence on the quality of mass spectra and sometimes completely suppress the * To whom correspondence should addressed. Tel.: +43/512/507-5125. Fax: +43/512/507-2965. E-mail:
[email protected]. (1) Flensburg, J.; Liminga, M. In The Proteomics Protocols Handbook; Walker, J. M., Ed.; Humana Press Inc.: Totowa, NJ, 2005; pp 325-340. (2) Krokhin, O. V.; Craig, R.; Spicer, V.; Ens, W.; Standing, K. G.; Beavis, R. C.; Wilkins, J. A. Mol. Cell. Proteomics 2004, 3, 908-919. (3) Wuhrer, M.; Hokke, C. H.; Deelder, A. M. Rapid Commun. Mass Spectrom. 2004, 18, 1741-1748.
8144 Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
signals.4,5 Due to the complexity of these samples, it is advantageous to separate the samples into several fractions. Size exclusion,6-8 ion exchange,9-12 and affinity chromatography13-16 can be used as techniques for the separation of protein mixtures. Separation of proteins by the above-mentioned techniques often requires mobile phases with higher ionic strengths. The separated protein fractions contain large amounts of salts, and therefore, purification steps (e.g., reversed-phase liquid chromatography) must be implemented in sample preparation prior to the determination of molecular masses by MALDI-TOF-MS.17-19 In addition, peptide mass fingerprinting plays a very important role in the identification of proteins. In this method, separated proteins are subjected to enzymatic cleavage and the peptide molecular masses determined by a proper mass spectrometric technique are submitted to database search program that identifies the protein within a selected protein database. Original protein samples often contain salts, and moreover, the digestion of proteins involves addition of salts; a suitable desalting procedure (4) Yao, J.; Scott, J. R.; Young, M. K.; Wilkins, C. L. J. Am. Soc. Mass Spectrom. 1998, 9, 805-813. (5) Keller, B. O.; Li, L. J. Am. Soc. Mass Spectrom. 2000, 11, 88-93. (6) Bublitz, R.; Kreusch, S.; Ditze, G.; Schulze, M.; Cumme, G. A.; Fischer, C.; Winter, A.; Hoppe, H.; Rhode, H. Proteomics 2006, 6, 3909-3917. (7) Salplachta, J.; Rehulka, P.; Chmelik, J. J. Mass Spectrom. 2004, 39, 13951401. (8) Ottens, M.; Houwing, J.; Van Hateren, S. H.; Van Baalen, T.; Van Der Wielen, L. A. M. Food Bioprod. Process. 2006, 84, 59-71. (9) Moure, F.; Rendueles, M.; Diaz, M. Bioprocess Biosyst. Eng. 2004, 27, 1724. (10) Shin, J.-H.; Krapfenbauer, K.; Lubec, G. Electrophoresis 2006, 27, 27992813. (11) Pepaj, M.; Wilson, S. R.; Novotna, K.; Lundanes, E.; Greibrokk, T. J. Chromatogr., A 2006, 1120, 132-141. (12) Yamakoshi, Y.; Hu, J. C. C.; Zhang, H.; Iwata, T.; Yamakoshi, F.; Simmer, J. P. Eur. J. Oral Sci. 2006, 114, 266-271. (13) Gong, Y.; Li, X.; Yang, B.; Ying, W.; Li, D.; Zhang, Y.; Dai, S.; Cai, Y.; Wang, J.; He, F.; Qian, X. J. Proteome Res. 2006, 5, 1379-1387. (14) Guerrier, L.; Thulasiraman, V.; Castagna, A.; Fortis, F.; Lin, S.; Lomas, L.; Righetti, P. G.; Boschetti, E. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2006, 833, 33-40. (15) Paunovic, I.; Schulin, R.; Nowack, B. J. Chromatogr., A 2005, 1100, 176184. (16) Bakry, R.; Gjerde, D.; Bonn, G. K. J. Proteome Res. 2006, 5, 1321-1331. (17) Hynek, R.; Svensson, B.; Jensen, O. N.; Barkholt, V.; Finnie, C. J. Proteome Res. 2007, 5, 3105-3113. (18) de Bont, J. M.; den Boer, M. L.; Reddingius, R. E.; Jansen, J.; Passier, M.; van Schaik, R. H. N.; Kros, J. M.; Sillevis Smitt, P. A. E.; Luider, T. H.; Pieters, R. Clin. Chem. 2006, 52, 1501-1509. (19) Linke, T.; Ross, A. C.; Harrison, E. H. J. Chromatogr., A 2006, 1123, 160169. 10.1021/ac0712392 CCC: $37.00
© 2007 American Chemical Society Published on Web 10/04/2007
is again necessary for high-quality mass spectra of enzymatic digests, as in the case of MS analysis of intact proteins.20,21 Several methods have been reported for sample purification prior to MS analysis. These include dialysis,22,23 ultrafiltration,24-26 size exclusion, affinity purification, and solid-phase extraction (SPE). Ultrafiltration and size exclusion spin columns do not provide a sufficient degree of desalting for MS. Dialysis allows a high degree of desalting, but it is difficult and expensive to automate. SPE with reversed-phase liquid chromatography (RP-LC) is the common strategy to desalt and concentrate protein and peptide samples. Mainly silica C18 is used for desalting; however, this does not allow the detection of small or hydrophilic peptides, or peptides altered in hydrophilicity such as phosphopeptides. Other chromatographic materials such as hydrophilic interaction, immobilized metal ion affinity chromatography,27 or porous graphite carbon (PGC)28 material have been proposed for purifying those categories of peptides prior to MS. PGC columns are normally used to purify carbohydrates and glycopeptides, but have recently been shown as an alternative or supplement to traditional RP-LC for separation of small and hydrophilic peptides prior to MALDITOF-MS analysis, nonetheless with lower sensitivity. The aim of this work is to develop and optimize the purification and identification procedures for peptides and proteins in a broad molecular mass range in chromatographic fractions containing high concentrations of salts based on MALDI-TOF-MS analyses of both the intact proteins and their tryptic digests. We have explored the use of C60-fullerenes covalently bound to silica as an alternative or supplement to RP microcolumns for desalting and preconcentration of protein and peptide mixtures. Several advantages of such materials are found in combination with MALDI-TOF-MS, including recovery or detection of small and hydrophilic peptides resulting in improved sequence coverage. EXPERIMENTAL SECTION Materials. C60-fullerene (g 99.5%) was purchased from MER Corp. (Tucson, AZ). Triethylamine (99.5%), (aminopropyl)trimethoxysilane (97.0%), 3-chloroperoxybenzoic acid (70-75%), R-cyano-4-hydroxycinnamic acid (HCCA; g99.0%), sinapinic acid (SA; g99.0%), trifluoroacetic acid (g99.0%), sodium sulfate anhydrous (99.0%), toluene (99.0%), tetrahydrofuran (THF; g99.9%), silica gel 60 (pore size 60 Å, 200-425 mesh, g98.5%), methanol (g 99.8%), sodium chromate tetrahydrate (99.0%), disodium hydrogen phosphate dihydrate (99.5%), orthophosphoric acid (85.0%), ammonium hydroxide (25.0%), sodium hydroxide (99.5%), acetonitrile (99.5%), insulin, ubiquitin (g90.0%), myoglobin (g95.0%), bradykinin (g98.0%), 6-hydroxyflavone (g98.0%), biochanin A (20) Xu, S.; Ye, M.; Xu, D.; Li, X.; Pan, C.; Zou, H. Anal. Chem. 2006, 78, 25932599. (21) Leite, J. F.; Hajivandi, M. R.; Diller, T.; Pope, R. M. Rapid Commun. Mass Spectrom. 2004, 18, 2953-2959. (22) Linnemayr, K.; Rizzi, A.; Josic, D.; Allmaier, G. Anal. Chim. Acta 1998, 372, 187-200. (23) Wu, Q.; Liu, C.; Smith, R. D. Rapid Commun. Mass Spectrom. 1996, 10, 835-838. (24) Tessier, B.; Harscoat-Schiavo, C.; Marc, I. J. Agric. Food Chem. 2006, 54, 3578-3584. (25) Chernokalskaya, E.; Kavonian, M. Am. Biotechnol. Lab. 2004, 22, 20-22. (26) Schratter, P. Methods Molecular Biology; Humana Press: Totowa, NJ, 2004, Vol. 244, pp 101-116. (27) Porath, J.; Carlsson, J.; Olsson, I.; Belfrage, G. Nature 1975, 258, 598599. (28) Chin, E. T.; Papac, D. I. Anal. Biochem. 1999, 273, 179-185.
(g97.0%), hesperetin (g95.0%), and naringenin (g95.0%) were purchased from Sigma-Aldrich (Vienna, Austria). Tryptic digest of bovine serum albumin (BSA) was purchased from Bruker (Bruker Daltonics, Bremen, Germany). Synthetic phosphopeptides: DpSEGRGpSGDPGK (M + H)+ ) 1321.45 Da and VYGKTpSHLR (M + H)+ ) 1140.56 Da from Bachem (Weil am Rhein, Germany). Kovasil 100A-5 (100 Å, 5 µm) silica gel was purchased from Zeochem AG (Uetikon, Switzerland). GromSIL 1000 Si (1000 Å, 5 µm) silica gel was obtained from Grom Analytik (Rottenburg-Hailfingen, Germany), and ProntoSil 300-5-Si (300 Å, 5 µm) was purchased from Bischoff Analysentechnik and Gera¨te GmbH (Leonberg, Germany). Water purified by a NanoPure-unit (Barnstead, Boston, MA) was used for all experiments. Synthesis of C60-Aminosilica. The synthesis of C60 bound to silica was carried out on three different silica materials having different pore sizes (Kovasil 100-5-Si (100 Å, 5 µm), ProntoSil 3005-Si (300 Å, 5 µm), and GromSil 1000-5-Si (1000 Å, 5 µm)). The silica sorbents were derivatized with an aminopropyl linker as described by Pesek et al.29 Two different functionalized fullerenes, namely, C60-fullerenoacetic acid30 and C60-epoxyfullerenes, were utilized.31 According to the values obtained by elemental analysis of aminosilica, the amount of amino groups were calculated (0.0032 mmol/m2). Three times molar excess of C60-fullerenoacetyl chloride (0.336 mmol) in 10 mL of dry THF and C60-epoxyfullerenes (0.336 mmol) in 150 mL of toluene were refluxed with 100 mg of aminosilica. The scheme of the preparation is demonstrated in Figure 1. The product of synthesis A was purified by washing and refluxing with THF, followed by boiling with a methanol/water solution (1:1) to hydrolyze the unreacted C60fullerenoacetyl chloride. The product was then finally washed with THF and dried. C60-epoxyfullerene bound to silica (synthesis B) was centrifuged and thoroughly washed with toluene, followed by removal of byproducts with a Soxhlet extractor for 12 h. The product was dried under vacuum. Sample Preparation for Solid-Phase Extraction. For all solid-phase extraction experiments, 5 mg of C60-aminosilica material was placed into 500-µL Eppendorf centrifuge tubes. The activation, equilibration, and washing steps were performed by shaking the suspension at 1400 rpm with a thermomixer (Eppendorf Thermomixer Comfort, Hamburg, Germany) for 3 min at room temperature. This was followed by centrifugation at 13 000 rpm (Eppendorf Centrifuge 5415D) at room temperature. The flowthrough, washing steps and elutions were investigated with MALDI-TOF/TOF-MS (Bruker Daltonics). SPE material from Waters, Oasis HLB (m-divinylbenzene and N-vinylpyrrolidone copolymer), and Sep-Pak (silica-based C18) as well as SPE material from Millipore (ZipTip silica-based C18) were used for the comparative study. The solid-phase extraction protocol for both C60-aminosilica and commercially available SPE sorbents was adopted according to Gilar et al.32 Solid-Phase Extraction of Peptides. Five milligrams of C60aminosilica was activated by suspending in 200 µL of acetonitrile (29) Buszewski, B.; Jezierska-Switala, M.; Kaliszan, R.; Wojtczak, A.; Albert, K.; Bachmann, S.; Matyska, M. T.; Pesek, J. J. Chromatographia 2001, 53, 204212. (30) Ito, H.; Tada, T.; Sudo, M.; Ishida, Y.; Hino, T.; Saigo, K. Org. Lett. 2003, 5, 2643-2645. (31) Balch, A. L.; Costa, D. A.; Noll, B. C.; Olmstead, M. M. J. Am. Chem. Soc. 1995, 117, 8926-8932. (32) Gilar, M.; Belenky, A.; Wang, B. H. J. Chromatogr., A 2001, 921, 3-13.
Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
8145
Figure 1. Reaction scheme showing the synthesis of fullerene-bonded silica, using (A) C60-fullerenoacetic acid and (B) C60-epoxyfullerene as starting material.
(ACN) in the presence of 0.1% trifluoroacetic acid (TFA), and the resultant mixture was ultrasonified for 5 min. After removing the supernatant, particles were equilibrated with 200 µL of PBS buffer. C60-aminosilica was loaded with 200 µL of three different peptide solutions, diluted in PBS, and incubated for 5 min at room temperature. The first solution was a mixture of insulin and bradykinin, concentrations ranging from 50 to 1000 µg/mL, used for quantitative analysis. The second was a peptide mixture containing angiotensin I (pI 6.99), angiotensin II (pI 6.70), substance P (pI 6.70), bombesin (pI 6.88), renin substrate (pI 5.11), ACTH clip 1-17 (pI 6.70), and ACTH clip 18-39 (pI 5.48) with a concentration of 50 µg/mL, for qualitative analysis with MALDITOF-MS. The third solution was a BSA tryptic digest, (100 µg/ mL) employed for data searching against Swiss-Prot using Mascot (Matrix Science Ltd.). The nonbound peptides were removed by washing the SPE material three times with 200 µL of PBS buffer followed by a desalting step with 200 µL of water. Finally, the peptides were eluted from SPE sorbent using 50 µL of 80% ACN in 0.1% TFA solution. The eluted peptides were analyzed with MALDI-TOFMS. Insulin and bradykinin were quantified with the RP-HPLC system from Bischoff modular HPLC instrument (Leonberg, Germany), equipped with a micropump (model 2250), a vacuum degasser (CSI 6150), and a photodiode array detector (Lamda 1010). The column, purchased from Sigma-Aldrich (4.6 mm × 100 mm), was packed with Hypersil-ODS silica material (end capped, 3 µm). Mc DAcq32 Data acquisition software (Bischoff) was used. All analyses were carried out at a mobile-phase flow rate of 1 mL/min. The peptides of interest were detected at 220 nm. Quantification was performed with the following mobile-phase compositions. For insulin, eluent A (80% 1/15 M KH2PO4 pH 2.1, 20% ACN) (v/v) and eluent B (60% 1/15 M KH2PO4 pH 2.1, 40% ACN) (v/v) were used for following linear gradient elution: from 35 to 100% B within 20 min. For the quantification of bradykinin, eluent A (99% H2O, 1% ACN, 0.1% TFA) (v/v) and eluent B (99% ACN, 1% H2O, 0.1% TFA) (v/v) were applied in the following gradient: from 15 to 100% B in 20 min. Solid-Phase Extraction of Phosphopeptides. The activation of C60-aminosilica was carried out first with 200 µL of 0.1% TFA 8146
Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
in ACN and then with 200 µL of 0.1% TFA in 1:1 ACN/water. In the third step, the material was equilibrated twice with 0.1% TFA. The activated and equilibrated C60-aminosilica was then incubated with 200 µL of phosphopeptides dissolved in PBS (50-1000 µg/ mL) for 5 min at 27 °C, followed by three washing steps with 5% ACN/0.1% TFA in water. The bound contents were eluted with 50 µL of 1:9:90, H3PO4/H2O/ACN. Phosphopeptides were quantified with the HP3D-CE system (Agilent Technologies, Waldbronn, Germany) equipped with a diode-array detector. Data processing was carried out with a HP3D-CE Chemstation (Rev. A.06.03, Agilent Technologies) software package. A polyimide-coated fusedsilica capillary, (50-µm i.d. and 360-µm o.d.) from Polymicro Technologies (Phoenix, AZ) was used for separation. The phosphopeptides were detected at a wavelength of 192 nm. All separations were performed by applying -30 kV at 22 °C. Phosphopeptides were injected by applying 50 mbar for 4 s. The capillaries with a total length of 68 cm and an effective length of 59.5 cm were conditioned by flushing with 0.1 M NaOH (for 20 min) followed by water (for 15 min). Finally, the conditioning procedure was completed by flushing with background electrolyte (BGE) for 20 min. Prior to each run, the capillary was purged (50 mbar) with the separation BGE for 5 min. BGE was prepared from 20 mM disodium hydrogen phosphate dihydrate in water. pH was adjusted to 2.5 with 85% orthophosphoric acid, and 0.001% HDB was added as EOF modifier to the electrolyte for co-EOF mode. Solid-Phase Extraction of Flavanoids. C60-aminosilica was activated with 2 mL of methanol and washed afterward with 1 mL of water. The sample was loaded with 40 µg/mL flavanoids standard dissolved in 0.1 M phosphate buffer, pH 2.4. After loading, three wash steps with 1 mL of water were carried out, followed by the elution with 1 mL of ACN. The separation and quantification was carried out using the capillary electrophoresis (CE) method according to Bachman et al.33 Hesperetin and naringenin were detected by UV detector at 214-nm wavelength whereas 6-hydroxyflavone and biochanin A were detected at 254 nm. BGE was prepared from sodium chromate tetrahydrate with a concentration of 1 mM by dissolving in water. pH was adjusted (33) Bachmann, S.; Bakry, R.; Huck, C. W.; Bonn, G. K. Electrophoresis 2007, 28, 799-805.
to 9.5 with ammonium hydroxide. All separations were performed by applying +30 kV at 22 °C. Solid-Phase Extraction of Proteins. Insulin, ubiquitin, and myoglobin were dissolved in PBS to get a protein mixture concentration of 100 µg/mL. The activation, loading, washing, and elution were carried out according to the protocol mentioned above for SPE of peptides. Instrumentation. Mass spectra were recorded on Ultraflex II MALDI-TOF/TOF-MS, (Bruker Daltonics), operated in reflector mode for the analysis of peptides and characterization of fullerene derivatives and in linear mode for the analysis of proteins. The 0.5-µL aliquots of the samples were placed onto a stainless steel target (MTP 384 target ground steel TF, Bruker Daltonics) with 0.5 µL of matrix (HCCA for peptide and SA for protein analysis). All spectra were recorded by summing 400 laser shots. The laser power was adjusted between 30 and 50% of its maximal intensity, using a 337-nm nitrogen laser having a pulse of 50 Hz. The instrument was calibrated externally using the peptide calibration standard II, purchased from Bruker. The Flex Analysis version 2.4 software packages provided by the manufacturer were used for data processing. NIR spectra were recorded on a single-beam polarization NIR Fourier transform spectrometer (FT-NIR; Bu¨chi, Flawil, Switzerland) equipped with a tungsten-halogen lamp and a temperated lead sulfide detector (30 °C) coupled to a horizontal sample desk for diffuse reflection measurements. Wavenumber ranges from 4000 to 10000 cm-1 (1000-2500 nm). The instrument offers a spectral resolution of 12 cm-1, an absolute wavelength accuracy of (2 cm-1, and a relative reproducibility of 0.5 cm-1. A light scattering cell (Hellma, QS 540.110, V ) 2800 µL) was used for scanning samples in reflectance mode. Chemometric software NirCal 4.21 (Bu¨chi) was used for creating a model; i.e., selection of spectra, wavelengths, mathematical pretreatment, and statistical analysis performing cluster analysis and principal component analysis. Spectra were randomly divided into a learning set (67%, C-set), i.e., calibration samples, and a test set (33%, V-set), i.e., samples for testing the calibration equation. The optimum number of factors used for the individual prediction was determined by cross-validation, and quality of cluster analysis was described in the Q-value (1 ) perfect calibration) calculated by the NirCal 4.21 software. RESULTS AND DISCUSSION Characteristics of the C60-Aminosilica Materials. The increases in carbon content of different synthetic approaches were investigated, with elemental analysis, to achieve maximum C60 functionalization on silica. In the first approach, the reaction of underivatized fullerenes with either silica (ProntoSil 300-5-Si) or aminosilica showed no increase in carbon content. This shows that the reaction had a very poor yield due to the lack of reactive sites on the fullerene. As a next step, reactive fullerene derivatives, namely, C60-epoxyfullerenes, were added to underivatized silica, which led to a carbon content of 6.1%. However, the highest carbon content was achieved when derivatized fullerenes, i.e., C60epoxyfullerenes and C60-fullerenoacetic acid, were reacted with aminosilica producing 17.54 and 11.3% carbon content, respectively. The results are summarized in Table 1. Comparing the C60 surface coverage calculated from elemental analysis, it can be seen that C60-epoxyfullerenes possess higher
Table 1. Elemental Analysis of Different Synthetic Approaches of C60-Silica reagents
N (%)
C (%)
H (%)
aminosilica fullerene + aminosilica epoxyfullerene + silica epoxyfullerene + aminosilica C60-fullerenoacetic acid + aminosilica
2.30 2.10 0 1.62 1.80
4.90 4.85 6.12 17.54 11.30
1.33 1.22 1.09 2.45 1.50
reactivity toward aminosilica than the counterpart C60-fullerenoacetid acid. C60-epoxyfullerenes having a C60 surface coverage of 2.9 mg of C60/m2 versus C60-fullerenoacetic acid with 1.2 mg of C60/m2. Parameters such as polarity and specific surface area of the sorbent both influence the retention of compounds. C60derivatized silica materials with different pore sizes (100, 300, and 1000 Å) were investigated for the SPE technology. Physical properties such as surface area, C60 coverage, and change in pore size before and after derivatization were measured. To investigate the change in surface area, a classical effective method named BET was utilized. There was ∼20% decrease in surface area for all silica after addition of the aminopropyl linker. On the other hand, an increase of 25-30% in surface area was observed after aminosilica was derivatized with fullerenes (increase of 15 m2/g for Kovasil 100 Å, 32.5 m2/g for Prontosil 300 Å, and 8.1 m2/g for Gromsil 1000 Å). Mercury intrusion porosimetry gave an insight into the change of average pore radius and volume. Considering these results, it can be observed that the pore volume was decreased immensely from 10 mm3/g to zero for Kovasil 100 and from 575 to 300 mm3/g for Prontosil 300. The pores of 100-Å silica are blocked to a great extent by C60-fullerene (70 Å). This observation was further confirmed by the decrease of average pore radius, namely, from 120 to 0 Å. In case of 300-Å silica, the volume is decreased by half, leading to the conclusion that pores are half filled with fullerenes. Table 2 summarizes the above-mentioned results. C60-epoxyfullerenes are chosen as starting material for different pore-sized silica derivatizations, due to their easy handling, less time-consuming synthesis, and cost efficiency. Characterization of the C60-Aminosilica Materials Using NIR. Near-infrared spectroscopy was applied to verify implemented derivatization steps of C60-fullerene samples. KubelkaMunk’s theory [f(R) ) (1 - R)2/2R, R ) I/I0)] was used to obtain information about the scattering and absorption coefficients of the samples integrated.34 In case of the fullerene-based materials, a constant amount of each sample was scanned 10 timessthe final spectra are presented as an average thereof. Absorption spectra in Figure 2A depict the recorded spectra of pure silica material (particle size 5 µm; pore size 100 Å) and the same silica material chemically modified. An intense additional band of the modified sample at 4878 and 6667 cm-1 could be clearly identified as an asymmetric N-H stretching vibration and a N-H stretching first overtone, respectively. After verification of the derivatized C60epoxyfullerenes with MALDI, the final products consisting of a combined silica-fullerene material were characterized. Spectra of C60-aminosilica (Figure 2 B) show several absorption bands that can be assigned to specific vibrational modes. On one hand, (34) Guthrie A. J.; Narayanaswamy, R.; Russell D. A. Analyst 1988, 113, 457461.
Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
8147
Figure 2. NIR absorption spectra of the synthetic steps of C60-aminosilica. (A) Synthesis of aminosilica; (B) synthesis of C60-aminosilica. Table 2. Characterization of Silica Gels Used in This Study (Data Acquired by the Manufacturers) and Characterization of the Aminopropyl Silica and the Fullerene Derivatives
name
specific surface area (m2/g)
average pore radius (µm)
average pore volume (mm3/g)
surface coverage (µmol/m2)
carbon content (%)
Kovasil 100 Kovasil 100-NH2 Kovasil 100-NH2-C60
305 250 265
120 Å
